Re: [PATCH v4 05/11] md/raid5: make sure max_sectors is not less than io_opt

["Yu Kuai" <yukuai@fnnas.com>]

Hi,

在 2026/1/13 13:06, Dan Carpenter 写道:
> Hi Yu,
>
> kernel test robot noticed the following build warnings:
>
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
>
> url:    https://github.com/intel-lab-lkp/linux/commits/Yu-Kuai/md-merge-mddev-has_superblock-into-mddev_flags/20260112-123233
> base:   linus/master
> patch link:    https://lore.kernel.org/r/20260112042857.2334264-6-yukuai%40fnnas.com
> patch subject: [PATCH v4 05/11] md/raid5: make sure max_sectors is not less than io_opt
> config: i386-randconfig-141-20260113 (https://download.01.org/0day-ci/archive/20260113/202601130531.LGfcZsa4-lkp@intel.com/config)
> compiler: gcc-14 (Debian 14.2.0-19) 14.2.0
> smatch version: v0.5.0-8985-g2614ff1a
>
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@intel.com>
> | Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
> | Closes: https://lore.kernel.org/r/202601130531.LGfcZsa4-lkp@intel.com/
>
> New smatch warnings:
> drivers/md/raid5.c:8100 raid5_run() warn: missing error code 'ret'
>
> vim +/ret +8100 drivers/md/raid5.c
>
> cc6167b4f3b3ca NeilBrown            2016-11-02  8064  	pr_info("md/raid:%s: raid level %d active with %d out of %d devices, algorithm %d\n",
> cc6167b4f3b3ca NeilBrown            2016-11-02  8065  		mdname(mddev), conf->level,
> ^1da177e4c3f41 Linus Torvalds       2005-04-16  8066  		mddev->raid_disks-mddev->degraded, mddev->raid_disks,
> e183eaedd53807 NeilBrown            2009-03-31  8067  		mddev->new_layout);
> ^1da177e4c3f41 Linus Torvalds       2005-04-16  8068
> ^1da177e4c3f41 Linus Torvalds       2005-04-16  8069  	print_raid5_conf(conf);
> ^1da177e4c3f41 Linus Torvalds       2005-04-16  8070
> fef9c61fdfabf9 NeilBrown            2009-03-31  8071  	if (conf->reshape_progress != MaxSector) {
> fef9c61fdfabf9 NeilBrown            2009-03-31  8072  		conf->reshape_safe = conf->reshape_progress;
> f67055780caac6 NeilBrown            2006-03-27  8073  		atomic_set(&conf->reshape_stripes, 0);
> f67055780caac6 NeilBrown            2006-03-27  8074  		clear_bit(MD_RECOVERY_SYNC, &mddev->recovery);
> f67055780caac6 NeilBrown            2006-03-27  8075  		clear_bit(MD_RECOVERY_CHECK, &mddev->recovery);
> f67055780caac6 NeilBrown            2006-03-27  8076  		set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery);
> ad39c08186f8a0 Yu Kuai              2024-02-01  8077  		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
> f67055780caac6 NeilBrown            2006-03-27  8078  	}
> f67055780caac6 NeilBrown            2006-03-27  8079
> ^1da177e4c3f41 Linus Torvalds       2005-04-16  8080  	/* Ok, everything is just fine now */
> a64c876fd35790 NeilBrown            2010-04-14  8081  	if (mddev->to_remove == &raid5_attrs_group)
> a64c876fd35790 NeilBrown            2010-04-14  8082  		mddev->to_remove = NULL;
> 00bcb4ac7ee7e5 NeilBrown            2010-06-01  8083  	else if (mddev->kobj.sd &&
> 00bcb4ac7ee7e5 NeilBrown            2010-06-01  8084  	    sysfs_create_group(&mddev->kobj, &raid5_attrs_group))
> cc6167b4f3b3ca NeilBrown            2016-11-02  8085  		pr_warn("raid5: failed to create sysfs attributes for %s\n",
> 5e55e2f5fc95b3 NeilBrown            2007-03-26  8086  			mdname(mddev));
> 4a5add49951e69 NeilBrown            2010-06-01  8087  	md_set_array_sectors(mddev, raid5_size(mddev, 0, 0));
> 7a5febe9ffeecd NeilBrown            2005-05-16  8088
> 176df894d79741 Christoph Hellwig    2024-03-03  8089  	if (!mddev_is_dm(mddev)) {
> f63f17350e5373 Christoph Hellwig    2024-03-03  8090  		ret = raid5_set_limits(mddev);
> f63f17350e5373 Christoph Hellwig    2024-03-03  8091  		if (ret)
> f63f17350e5373 Christoph Hellwig    2024-03-03  8092  			goto abort;
> 9f7c2220017771 NeilBrown            2010-07-26  8093  	}
> 23032a0eb97c8e Raz Ben-Jehuda(caro  2006-12-10  8094)
> 585d578974395f Yu Kuai              2026-01-12  8095  	ret = raid5_create_ctx_pool(conf);
> 585d578974395f Yu Kuai              2026-01-12  8096  	if (ret)
> 01fce9e38c0e92 Yu Kuai              2026-01-12  8097  		goto abort;
> 01fce9e38c0e92 Yu Kuai              2026-01-12  8098
> 845b9e229fe071 Artur Paszkiewicz    2017-04-04  8099  	if (log_init(conf, journal_dev, raid5_has_ppl(conf)))
> 5aabf7c49d9ebe Song Liu             2016-11-17 @8100  		goto abort;
>
> Presumably we should propagate the error code from log_init()?

Yes we should, but this problem looks already exist before this patch.

>
> 5c7e81c3de9eb3 Shaohua Li           2015-08-13  8101
> ^1da177e4c3f41 Linus Torvalds       2005-04-16  8102  	return 0;
> ^1da177e4c3f41 Linus Torvalds       2005-04-16  8103  abort:
> 7eb8ff02c1df27 Li Lingfeng          2023-08-03  8104  	md_unregister_thread(mddev, &mddev->thread);
> ^1da177e4c3f41 Linus Torvalds       2005-04-16  8105  	print_raid5_conf(conf);
> 95fc17aac45300 Dan Williams         2009-07-31  8106  	free_conf(conf);
> ^1da177e4c3f41 Linus Torvalds       2005-04-16  8107  	mddev->private = NULL;
> cc6167b4f3b3ca NeilBrown            2016-11-02  8108  	pr_warn("md/raid:%s: failed to run raid set.\n", mdname(mddev));
> f63f17350e5373 Christoph Hellwig    2024-03-03  8109  	return ret;
> ^1da177e4c3f41 Linus Torvalds       2005-04-16  8110  }
>
-- 
Thansk,
Kuai